- From: David Carlisle <davidc@nag.co.uk>
- Date: Mon, 25 Apr 2011 23:15:30 +0100
- To: Fraser Goffin <goffinf@gmail.com>
- CC: www-math@w3.org
On 22/04/2011 16:33, Fraser Goffin wrote:
If you modify mathml3-strict-content.xsd definition of ContExp to be
more like the RelaxNG version then the example I posted validates
<xs:group name="ContExp">
<xs:choice>
<!--Ambiguous content model altered (ContExp)-->
<xs:element ref="m:apply"/>
<xs:element ref="m:bind"/>
<xs:element ref="m:ci"/>
<xs:element ref="m:cn"/>
<xs:element ref="m:cs"/>
adding the last line adds back cs to the content model of apply.
(looks like cerror, cbytes and share need adding at the same place)
The comment at the start of the choice is a warning from trang that it
"altered" the content model, sorry for not noticing quite how much it
was altered.
I need to do a few more checks as to where exactly in the conversion
things broke down, but adding these elements back to the ContExp group
seems like a correct fix in the meantime...
David
Received on Monday, 25 April 2011 22:15:48 UTC